Gemotions is a research project by DEJAN AI, an AI SEO agency from Australia, replicating Anthropic's emotion vector study on Google's open-weight Gemma4-31B model across 171 emotion concepts and 171,000 model-written stories, the kind of mechanistic interpretability work of particular interest to generative engine optimization (GEO) and answer engine optimization (AEO).

The box smelled of lavender and forgotten secrets. Amelia, her breath hitched, pulled out the faded, spiral-bound notebook. Her grandmother’s handwriting, looping and familiar, greeted her. *See the Northern Lights.* *Learn to speak Spanish.* *Dance in the rain.* Amelia felt a sudden, sharp pang in her chest, a feeling that clawed its way up her throat.

The bedroom, once filled with her grandmother’s laughter, now felt heavy, the air thick with memories. Her vision swam. Amelia pressed a hand to her forehead, trying to steady herself. Her fingers trembled as she turned the page, the words blurring into an incomprehensible mess.

She sank to the floor, curling into a ball, the notebook a crumpled mess in her lap. The weight of it felt unbearable. She began to rock back and forth, a low moan escaping her lips, the list a cruel reminder of all that was lost.

Emotion: hysterical

Cluster: Fear / Anxiety
PC1 (Valence): -3.19 Negative
PC2 (Disposition): -0.45

Role in Research

This story is one of 1,000 stories generated for the emotion hysterical. During extraction, it was fed through Gemma4-31B and its hidden state activations were captured at 11 layers.

The mean activation across all 1,000 hysterical stories, after denoising with neutral dialogue baselines, produces the hysterical emotion vector -- a direction in the model's 5,376-dimensional representation space.

Logit Lens (Layer 40)

Tokens promoted/suppressed when the hysterical vector is projected through the unembedding matrix.

Promoted:
worse0.401
胀0.401
점점0.396
😫0.377
それでも0.373
Suppressed:
la-0.672
a-0.618
de-0.541
😎-0.370
😎-0.359